All articles
Jul 23, 2026·5 min read·Guide

Scam Guard: Stopping Compromised-Account Raids in Discord

The scam wave that hits every growing server comes from hacked accounts mass-pinging @everyone and flooding channels with a malicious link. Scam Guard catches that behaviour in seconds — even when the payload is an image no word filter can read.

Almost every Discord community eventually meets the same attack: a member's account gets compromised, and within seconds it blasts an @everyone ping and drops the same "free Nitro" link into ten channels at once. By the time a human moderator sees it, hundreds of people already have. Scam Guard is built for exactly this moment.

Why a word filter isn't enough

Word filters match text. Modern scam raids increasingly ship the link inside an image — a screenshot, a QR code, a fake giveaway banner — so there's no text to match. Scam Guard doesn't look at what is posted; it looks at how. Mass-mentioning everyone and posting the same thing across many channels in a few seconds is behaviour a real person almost never does, and that's the signal it acts on.

What it watches for

Two patterns give a compromised account away:

  • Mass mentions — a single message that pings @everyone or @here (or a pile of user/role mentions) from an account that shouldn't.
  • Multi-channel flooding — the same or near-identical message posted into several channels within a very short window.

Either pattern trips the guard.

What it does when it fires

When Scam Guard catches a raid pattern it takes the action you configured — ban, kick, timeout or message removal — immediately, then logs it so your team has a record. The point is speed: an automatic response that lands in the seconds before the damage spreads, instead of a manual clean-up afterwards.

It protects the raider too

Because these are usually hacked accounts, not malicious members, the right response is to stop the account fast and let the real owner recover it. Ban-and-appeal or a timeout does that cleanly — the account stops spamming, your members stay safe, and nobody loses their real identity permanently.

Layer it with the rest

Scam Guard sits alongside CloudMod's other moderation plugins — the word filter for text, the honeypot for bots that auto-post everywhere, the account-age filter for throwaway joins, and network-wide bans so one action covers every server in your network. Together they cover the raid surface that a single tool can't.

Getting-started tip

Turn Scam Guard on with a timeout action first and watch your moderation log for a few days. Once you trust what it's catching, switch the action to ban for the fastest possible stop. Give the bot Ban Members, Kick Members, Moderate Members and Manage Messages so every action option is available.

Try CloudMod for free

Setup in under 10 minutes. No credit card.

Get started